PU.1 regulates RUNX3 expression. (A) BM cells from wild-type (wt) and PU.1cKO mice were cultured for 3 d in the presence of GM-CSF and TGF-β. CD11c⁺MHCII⁺ cells were sorted and levels of Runx3 were measured by quantitative real-time PCR. Graphs indicate the mean Runx3 expression ± SD relative to Hprt. ***, P < 0.001 (unpaired Student’s t test) for the indicated comparison. Data are representative of three independent experiments. (B) BM cells from PU.1cKO were transduced with either the empty vector (blue), or PU.1 (red)-expressing retroviruses (RV). RUNX3 expression in CD11c⁺MHCII⁺ DCs was measured by flow cytometry and compared with the isotype control (black line). Graph shows the geometric MFI ± SD of RUNX3 in transduced DCs with the indicated vector. ***, P < 0.001 (unpaired Student’s t test) for the indicated comparison. (C) Mean numbers of CD11c⁺MHCII⁺DEC205⁺EPCAM⁺ LCs ± SD generated after transduction of PU.1cKO BM with either empty vector (blue) or PU.1 (red). Data in B and C are representative of three independent experiments. (D) PU.1 binding at the Runx3 locus as defined by ChIPseq analysis of BM derived macrophages. The PU.1 ChIPseq data have been previously described (Ghisletti et al., 2010). The PU.1 peaks that were subjected to further analysis in LCs are labeled A and B. Sequence below highlights additional putative PU.1 binding site (black boxes) found in the promoter of Runx3. Primers used for ChIP PCR are highlighted in gray. (E) Wild-type BM cells were cultured in GM-CSF ± TGF-β for 3 d. Binding of PU.1 to the regions labeled as A and B in D was analyzed by ChIP and enrichment was calculated by quantitative real time PCR as the mean fold of enrichment ± SD (compared with IgG control). Results are representative of two independent experiments.